Automatic speech recognition using articulatory features from subject-independent acoustic-to-articulatory inversion

• Published in: The Journal of the Acoustical Society of America Vol. 130; no. 4; pp. EL251 - EL257
• Main Authors: Ghosh, Prasanta Kumar, Narayanan, Shrikanth
• Format: Journal Article
• Language: English
• Published: Melville, NY Acoustical Society of America 01.10.2011; American Institute of Physics
• Subjects: Acoustic Phonetics; Applied sciences; Articulatory Phonetics; Automatic Speaker Recognition; Automation; Biological and medical sciences; Ear and associated structures. Auditory pathways and centers. Hearing. Vocal organ. Phonation. Sound production. Echolocation; English; Exact sciences and technology; Female; Fundamental and applied biological sciences. Psychology; Humans; Information, signal and communications theory; Jasa Express Letters; Jaw - anatomy & histology; Jaw - physiology; Linear Models; Male; Models, Theoretical; Mouth - anatomy & histology; Mouth - physiology; Phonation; Phonetics; Signal processing; Speech Acoustics; Speech processing; Speech Production Measurement; Speech Recognition Software; Telecommunications and information theory; Vertebrates: nervous system and sense organs; Voice Recognition; Vocal tract; Speech articulation; Verbal production; Inverse problem
• ISSN: 0001-4966; 1520-8524; 1520-8524
• DOI: 10.1121/1.3634122

An automatic speech recognition approach is presented which uses articulatory features estimated by a subject-independent acoustic-to-articulatory inversion. The inversion allows estimation of articulatory features from any talker's speech acoustics using only an exemplary subject's articulatory-to-acoustic map. Results are reported on a broad class phonetic classification experiment on speech from English talkers using data from three distinct English talkers as exemplars for inversion. Results indicate that the inclusion of the articulatory information improves classification accuracy but the improvement is more significant when the speaking style of the exemplar and the talker are matched compared to when they are mismatched.